Effective use of pneumococcal and influenza vaccines will require more precise definition of high risk (HR) populations, particularly among the elderly, and identification of appropriate sites for vaccines delivery. The proposed studies will use hospital discharge data from 14 hospitals participating in the Shenandoah Professional Standards Review Foundation (SPSRF) from July 1, 1978 through June 30, 1985. These data represent hospital experience for 72,000 Medicare enrollees. To test the hypothesis that previous hospital care may be a useful marker for identifying HR individuals, person-oriented files of all SPSRF patients hospitalized with pneumonia in 1982 will be created which include all admissions within the previous 3 to 5 years. Similar files will be created for persons hospitalized with pneumonia and influenza (P&I) and acute and chronic respiratory conditions (ACRC) during the 1983 influenza epidemic. The files will be examined historically to determine the proportion of patients with previous hospital care, analyzing by age/sex/risk status and outcome. To test the hypothesis that hospital-based pneumococcal and influenza immunization of discharged patients would prevent many hospital readmissions and deaths, cohort analysis will determine the probability of readmission within 3 to 5 years with pneumonia, or within 2 1/2 years with influenza-related P&I or ACRC. The preventive effect of immunizing discharged patients will be expressed as the number of persons who would have to be immunized on discharge to prevent each readmission and/or death due to preventable pneumococcal pneumonia or influenza. The studies are expected to show that a substantial proportion of preventable pneumonia and influenza occurs in patients with previous hospital care. Immunization of relatively few discharged patients, especially those discharged with HR conditions, could prevent each readmission and/or death due to these two diseases. The studies will have importance for (1) defining HR populations, (2) identifying sites for immunization programs, (3) justifying clinical trials of hospital-based immunization, (4) modifying federal and non-federal policies for vaccine use, and (5) assessing the cost-effectiveness and cost benefits of federal and hospital immunization programs. They will suggest hospital-based immunization as a new approach to more effective prevention of pneumococcal pneumonia and influenza among elderly persons.